Human cognition and behavior depend on the proper assembly and maintenance of neural circuits and genes that imperil these processes are increasingly linked to autism, schizophrenia and other neurological disorders. Neural activity and genetic programs contribute to different aspects of neural circuit development and architecture. In some sensory circuits, such as vision, hearing and touch, neurons with related response properties are organized in a stereotypic manner to form maps of sensory information. In these systems, genes and neural activity play complementary roles in forming and maintaining these maps. Less is known about the developmental principles that help to establish finer scale neural connectivity, particularly in processing regions that lack obvious spatial maps. Here, we propose to examine the impact of neural activity on circuits involved in the sense of smell. In particular, we will study the circuits formed by the mitral and tufted (MT) neurons, which receive inputs from defined groups of sensory neurons and transmit this information to multiple functionally distinct cortical processing centers responsible for olfactory learning and innate behaviors such as attraction or fear. MT neurons are important to study for two reasons. First, we have developed new viral tracing and neuron reconstruction techniques to label sets of sister MT neurons that respond to the same odors in the same animal and map their projections in three dimensions. This affords us a sensitive method to investigate fine scale wiring mechanisms. Second, individual MT neurons innervate multiple cortical targets, which each seems to posses a characteristic architecture. This allows us to simultaneously assess the impact of activity on multiple circuit architectures that are formed by different branches of the same MT axon. The central hypothesis of this proposal is that activity in MT neurons differentially regulates distinct aspects of olfactory circuit architecture. To test this hypothesi we propose three specific aims. First, we will determine whether projections of sister mitral and tufted (MT) neurons are stereotyped in cortical processing centers involved in innate behaviors. Second, we will synaptically silence olfactory sensory neurons and identify the features of olfactory bulb and cortical circuits that depend on sensory input. Third, we will synaptically silence MT neurons and identify aspects of local and cortical circuit architecture that depend on MT activity. Results of these studies will identify unknown activity-dependent mechanisms of olfactory neural circuit formation and predict which features of circuit architecture are likely tobe controlled by genetic programs. Defining these circuits with high resolution will facilitate future studies to identify genes implicated in building specific neural circuits and to assess the functional consequence of mutations linked to human neurodevelopmental or neurodegenerative diseases.